Human Reproduction vol.15 no.2 pp.
237–238, 2000
OPINION
Reclassification of azoospermia: the time has come?
Khaldoun Sharif
Assisted Conception Services, Birmingham Women’s Hospital,
Birmingham B15 2TG, UK
This opinion was previously published on Webtrack 93,
November 4, 1999
Azoospermia (the absence of spermatozoa from the ejaculate)
is not uncommon, and is present in ~5% of all investigated
infertile couples (Irvine, 1998), and in 10–20% of infertile
men with abnormal seminal fluid analysis (Stanwell-Smith and
Hendry, 1984). Over the past few years there has been
much renewed interest in the condition because of increased
understanding of its genetic basis and aetiology (Mak and
Jarvi, 1996) as well as the availability of surgical sperm
retrieval methods and intracytoplasmic sperm injection (ICSI)
for the treatment of the resulting infertility (Palermo et al.,
1999). As more advances are being made and publicised,
clinicians of different disciplines will be dealing with more
azoospermic patients than any other time in the past. It is,
therefore, both desirable and timely that there is a clear, logical
and clinically-orientated classification of azoospermia. This
classification should place the different causes of azoospermia
into distinct categories which (as much as possible) should be
comprehensive and have common aetiology, presentation,
prognosis and treatment. This would lend itself to a clearer
clinical thought process, and consequently a more correct
diagnosis and proper management. It is the opinion of this
author that the currently used traditional classification of
azoospermia falls short of these goals, and a different classi-
fication which better serves these purposes is suggested.
Traditionally azoospermia is classified into ‘obstructive’ and
‘non-obstructive’ (Prins et al., 1999). In obstructive cases
spermatogenesis is normal but there is obstruction in the
seminal ducts, while in non-obstructive cases there is deficient
(or absent) spermatogenesis. At the face of it this classification
looks descriptive and clear and, indeed, has been used for a
long time. But where do cases of azoospermia due to retrograde
ejaculation fit? Definitely there is normal spermatogenesis
but no real obstruction. Also, cases of azoospermia due to
hypogonadotrophic hypogonadism and those due to Sertoli
cell-only syndrome are placed in the ‘non-obstructive category’
although they have totally different aetiologies, treatments and
prognoses. And what about the patient who has had a vasectomy
but also later developed testicular dysfunction following
chemotherapy? Can we say that he has both obstructive and
non-obstructive azoospermia? Surely, that will confuse rather
than clarify the issue. The traditional classification is perhaps
© European Society of Human Reproduction and Embryology
simplistic (rather than simple), and forces divergent causes to
sit together, rather uncomfortably, in the same category, and
is a recipe for potential misdiagnosis and mismanagement.
It is a truism that one needs to appreciate the normal in
order to understand the abnormal. Normally, the hormones of
the hypothalamic–pituitary axis stimulate the testis, the testis
produces spermatozoa which travel to the exterior through
patent and properly functioning seminal ducts. Logically,
therefore, azoospermia could be divided into causes due
to deficient hormonal stimulation of the testis, testicular
dysfunction, and seminal ducts obstruction or dysfunction;
pre-testicular, testicular, and post-testicular causes respectively.
Indeed, a similar classification of male factor infertility in
general has been previously described in some textbooks and
papers (Berkow and Fletcher, 1992; Mak and Jarvi, 1996).
However, almost all current clinical and research papers on
azoospermia use the ‘obstructive/non-obstructive’ classifica-
tion with its aforementioned drawbacks. The following
classification should provide a more logical framework.
Pre-testicular azoospermia
This includes all cases of hypogonadotrophic hypogonadism,
whether congenital (Kallmann’s syndrome), acquired (trauma,
tumour), or idiopathic. These require pituitary assessment (both
endocrinological and radiological) and respond very well to
hormone replacement therapy (Finkel et al., 1985). After
successful spermatogenesis has been established, pregnancy
occurs without the aid of assisted conception.
Testicular azoospermia
This includes testicular disorders and may be congenital
(Klinefelter’s syndrome, Y-deletion), acquired (radiotherapy,
chemotherapy, torsion, mumps orchitis), or developmental
(testicular maldescent). Treatment of testicular causes require
attempt at surgical sperm retrieval, after appropriate genetic
screening. Spermatozoa are retrieved in only ~50% of these
cases (Gil-Salom et al., 1998). Prognosis, therefore, should
be guarded.
Post-testicular azoospermia
These cases are due to either ductal obstruction or dysfunction
(retrograde ejaculation). Obstruction could be congenital
(congenital bilateral absence of the vas deferens or CBAVD)
or acquired (post-surgery or infection). Acquired obstructive
237
K.Sharif

cases could be amenable to surgical correction. Those which

are unsuccessful, or unsuitable, as well as cases of CBAVD
will require surgical sperm retrieval for ICSI. Screening for
cystic fibrosis mutations should be undertaken in men with
CBAVD (Donat et al., 1997). Cases of retrograde ejaculation
may respond to sympathomimetics. Failing that, spermatozoa
could be retrieved from alkalinized post-ejaculation urine and
used for insemination or in-vitro fertilization/ICSI depending
on the available number and quality. Spermatozoa could be
obtained in almost all cases of post-testicular origin (Safran
et al., 1998).
Concern has been raised about the unchecked speed of
development and widespread application of treatments for
azoospermia (Steele et al., 1999). Also, infertile male patients
have sued clinicians for failing to diagnose lesions where an
initial accurate diagnosis and simpler proper treatment could
perhaps have avoided the use of expensive and invasive
assisted reproductive techniques, albeit successfully (Jequier,
1997). There is a timely need for a clear and logical
classification of azoospermia that is simple, clinically-
orientated, inclusive of all causes, and lends itself to therapeutic
strategies. The ‘pre-testicular, testicular, and post-testicular’
classification seems to fulfil all these aims.

References
Berkow, R. and Fletcher, A.J. (1992) The Merck Manual of Diagnosis and
Therapy. Merck Sharp & Dohme Research Laboratories, Rahway, NJ, USA.
Donat, R., McNeill, A.S., Fitzpatrick, D.R. et al. (1997) The incidence of
cystic fibrosis gene mutations in patients with congenital bilateral absence
of the vas deferens in Scotland. Br. J. Urol., 79, 74–77.
Finkel, D.M., Phillips, J.L. and Snyder, P.J. (1985) Stimulation of
spermatogenesis by gonadotropins in men with hypogonadotropic
hypogonadism. N. Engl. J. Med., 313, 651–655.
Gil-Salom, M., Romero, J., Minguez, Y. et al. (1998) Testicular sperm
extraction and intracytoplasmic sperm injection: a chance of fertility in
nonobstructive azoospermia. J. Urol., 160, 2063–2067.
Irvine, D.S. (1998) Epidemiology and aetiology of male infertility. Hum.
Reprod., 13 (Suppl. 1), 33–44.
Jequier, A. (1997) Clinical assessment of male infertility in the era of
intracytoplasmic sperm injection Baillière’s Clin. Obstet. Gynaecol., 11,
617–639.
Mak, V. and Jarvi, K.A. (1996) The genetics of male infertility. J. Urol., 156,
1245–1256 (discussion 1256–1247).
Palermo, G.D., Schlegel, P.N., Hariprashad, J.J. et al. (1999) Fertilization and
pregnancy outcome with intracytoplasmic sperm injection for azoospermic
men. Hum. Reprod., 14, 741–748.
Prins, G.S., Dolgina, R., Studney, P. et al. (1999) Quality of cryopreserved
testicular sperm in patients with obstructive and nonobstructive azoospermia.
J. Urol., 161, 1504–1508.
Safran, A., Reubinoff, B.E., Porat-Katz, A. et al. (1998) Assisted reproduction
for the treatment of azoospermia. Hum. Reprod., 13 (Suppl. 4), 47–60.
Stanwell-Smith, R.E. and Hendry, W.F. (1984) The prognosis of male
subfertility: a survey of 1025 men referred to a fertility clinic. Br. J. Urol.,
56, 422–428.
Steele, E.K., Lewis, S.E. and McClure, N. (1999) Science versus clinical
adventurism in treatment of azoospermia. Lancet, 353, 516–517.

238
 EDITOR’S CORNER

Surgical sperm retrieval: what not to do

Khaldoun Sharif, F.R.C.O.G., and Samer Ghunaim, M.Sc.
Istishari Fertility Centre, Istishari Hospital, Amman, Jordan

Surgical sperm retrieval has revolutionized the treatment of azoospermia. However, the ease with which it could be
performed meant that it is perhaps being used where not indicated. Here are the potential pitfalls to be avoided.
(Fertil Steril! 2008;89:17–9. "2008 by American Society for Reproductive Medicine.)

Azoospermia (the absence of sperm from the ejaculate) is spermic men, by definition, have sperm in the ejaculate,
present in 5% of all investigated infertile couples (1). Origi- which could be used for ICSI without the need for SSR.
nally it was thought to be an almost untreatable form of male
There have been suggestions that SSR for ICSI should be
infertility (2). More recently, with the advent of testicular/
performed in men with very low number of sperm in the ejac-
epididymal surgical sperm retrieval (SSR) and intracytoplas-
ulate with high DNA damage, based on the observation that
mic sperm injection (ICSI), many cases of infertility due to
the testis is rich in antioxidant enzymes, which may reduce
azoospermia are now successfully treated (3). However,
DNA damage (6). However, SSR is not without complica-
with the current wide availability of ICSI and the relative
tions, and there is no evidence to support performing it for
ease with which SSR could be performed by clinicians of
ICSI in men with sperm in the ejaculate, which is the view
different disciplines, perhaps some infertile couples are
endorsed by the ASRM Practice Committee (7).
undergoing SSR/ICSI treatment where it is not indicated or
in a suboptimal way. It is therefore necessary to have clear
clinical guidance, not just advising what should be done DO NOT DO SSR IN AZOOPSERMIC MEN WITH
(which is what most monograms and surgical texts tend to HYPOGONADOTROPHIC HYPOGONADISM
give) but also and as important, what should not be done.
About 1% of azoopsermic men have hypogonadotrophic
Here we present a number of ‘‘what-not-to-do’’ recommenda-
hypogonadism, and it is the rarity of this diagnosis that makes
tions in cases of SSR/ICSI. Although many of them are
the error in managing it more likely to occur. In these patients
apparently obvious, we have seen too many cases where
there will be clinical symptoms and signs of hypogonadism,
some of these practices were done previously, and therefore
low T, and low FSH. An SSR is not appropriate in these men,
believe that such a reminder is likely to be helpful.
who require proper assessment and hormone replacement
therapy (HT). This consists of T when fertility is not desired
and gonadotropins when it is desired. With proper treatment,
DO NOT DO SSR UNLESS IT IS TRUE AZOOSPERMIA normal spermatogenesis is induced in almost all patients
(3, 8–10). This may take more than 1 year, but spontaneous
Because the routine seminal fluid analysis is a highly subjec-
pregnancy usually follows in many cases. However, despite
tive test (4), it is not unusual for a specimen that is reported as
treatment some patients will not produce adequate sperm
azoospermic to contain a few sperm (cryptozoopsermia) after
numbers or quality to induce natural pregnancy, or the couple
centrifugation (pelleting) (5). In a study of 140 patients who
might simply not be willing to wait for the year or so it takes
were found to be azoospermic on routine semen analysis,
to reach that stage. In such cases, and after appropriate coun-
sperm were found in 20% of them on sperm pelleting. This
seling, one could offer assisted reproduction, but again with
was found in patients previously labeled as either being
ejaculated sperm rather than SSR.
obstructive or nonobstructive azoospermic, and irrespective
of the serum FSH level or findings on previous testicular
biopsy (5). Sperm pelleting, therefore, should be performed DO NOT DO SSR AS A FIRST LINE IN AZOOPSERMIC MEN
on semen samples from all seemingly azoospermic patients WITH RETROGRADE EJACULATION
to verify the diagnosis before considering SSR. Cryptozoo-
This diagnosis should be suspected in an azoospermic man
with low volume (<1 mL) ejaculate, and confirmed by find-
Received June 8, 2007; revised and accepted August 31, 2007.
ing sperm on examining postejaculation urine (11). The first
Reprint requests: Khaldoun Sharif, F.R.C.O.G., Istishari Fertility Centre,
Istishari Hospital, Alkindi Street, Wadi saqra, Amman, Jordan (FAX: line of treatment is a trial of medical therapy with a-symap-
9626-500-1006; E-mail: k.sharif@mac.com). thomymetics to induce antegrade ejaculation, and if this does

0015-0282/08/$34.00 Fertility and Sterility# Vol. 89, No. 1, January 2008 17

doi:10.1016/j.fertnstert.2007.08.075 Copyright ª2008 American Society for Reproductive Medicine, Published by Elsevier Inc.
not work, sperm could be retrieved from suitably prepared
postejaculatory urine for use with assisted conception (11).
The SSR is a last resort option here, as the alternatives are
less complex, highly successful, and less invasive.
DO NOT DO SSR AND ICSI WITHOUT FIRST DOING
APPROPRIATE GENETIC TESTING AND COUNSELING
The incidence of both structural and numerical chromosomal
abnormalities is increased in azoospermic patients. Genetic
testing, coupled with appropriate counseling, should be
performed to detect these abnormalities as this will aid in
the diagnosis, guide the need for preimplantation genetic
diagnosis, and help to quantify the genetic risk to the poten-
tial offspring (12).
Patients with nonobstructive azoospermia have up to 15%
incidence of numerical chromosomal abnormalities, mainly
Klinefelter’s syndrome (47, XXY) and occasionally balanced
translocations. The detection of chromosomal abnormalities
should lead to relevant genetic counseling and, in appropriate
cases, the recommendation for preimplantation genetic diag-
nosis to avoid replacing embryos with unbalanced transloca-
tions (13).
In addition to standard karyotyping, these patients should
undergo screening for microdeletions in the long arm of the
Y chromosome (Yq11), as these have been detected in
10%–15% of cases. Potentially, male children of these men
will carry the same microdeletion and exhibit the same
pattern of infertility in adulthood (14).
In addition, testing for cystic fibrosis transmembrane con-
ductance regulator gene mutations should be performed in
azoospermic men with congenital bilateral absence of the
vas, seminal vesicle abnormalities, or low volume acidic
ejaculate. These mutations are present in almost all patients
with congenital bilateral absence of the vas and other
Wolffian anomalies, but in some cases they are not detectable
on routine testing (15). Female partners of men with congen-
ital bilateral absence of the vas should be offered testing for
cystic fibrosis transmembrane conductance regulator to
assess the risk of producing babies with cystic fibrosis and
prevent it through preimplantation genetic diagnosis if both
partners carry the mutation (16).
It is noteworthy that the available data do not suggest an
increase in the incidence of major birth defects after the
use of SSR and ICSI (17).
DO NOT DO SSR UNLESS FACILITIES ARE AVAILABLE TO
EXAMINE THE WET PREPARATION
Diagnostic SSR is done to determine whether there are sperm
in the testis. This could be ascertained by two means: direct
examination of a fresh wet preparation or histologic examina-
tion of fixed tissue. If there is no facility or expertise available
to immediately examine a wet preparation, the only thing left
to rely on will be the histologic examination. However, sperm
are found in wet preparations more often than suggested by
18 Sharif and Ghunaim Pitfalls in surgical sperm retrieval
the histologic examination (18, 19). Therefore, complete
reliance on the histologic examination will label some
patients inaccurately as having no testicular sperm and
deny them the chance of a potentially successful treatment.
DO NOT RELY ON A SINGLE TESTICULAR BIOPSY OR
UNILATERAL TESTICULAR BIOPSIES
In cases of nonobstructive azoospermia a single testicular
biopsy is not adequate to rule out the presence of sperm, as
spermatogenesis may be present only in certain foci of
seminiferous tubules, and multiple testicular sampling will
lead to sperm recovery in up to 60% of cases (20, 21).
Also, biopsies from both testes are necessary, as it has been
shown that in 20% of patients focal spermatogenesis would
have been missed after unilateral biopsy (22).
DO NOT DO DIAGNOSTIC SSR UNLESS SPERM
CRYOPRESERVATION FACILITIES ARE AVAILABLE
The aim of a diagnostic SSR is to determine whether there are
sperm so they could be used later on with ICSI. Ideally, this
diagnostic procedure should be done in a unit with sperm
cryopreservation facilities, to avoid having to put the patient
unnecessarily through another SSR at the time of ICSI.
Cryopreserved testicular sperm are as good as fresh testicular
sperm in their fertilizing ability and pregnancy rates (PR)
(23).
DO NOT REPEAT SSR UNLESS A SUITABLE PERIOD OF
TIME HAS PASSED TO ALLOW RECOVERY OF THE TESTIS
It is well recognized that SSR leads to transient changes in the
testis, probably resulting from hematoma formation and
inflammation (24). These changes can affect the spermato-
genic efficiency of the testis, and it has been demonstrated
that after a positive SSR (where sperm have been found),
a repeat procedure is less likely to yield sperm if performed
sooner (less than 3–6 months after the initial SSR) rather
than later (24, 25).
DO NOT DO SSR IN POST-TESTICULAR (OBSTRUCTIVE)
AZOOSPERMIA WITHOUT FIRST CONSIDERING SURGICAL
CORRECTION
The commonest cause of post-testicular azoospermia is pre-
vious surgery (vasectomy). There are many advantages to
surgical correction over SSR/ICSI. If successful, the surgery
will allow the couple to achieve many pregnancies, whereas
a new ICSI cycle is needed for each attempt. Second, surgery
avoids the high chance of multiple pregnancy associated with
ICSI and multiple embryo transfer. Third, performing correc-
tive surgery may be more cost effective than sperm retrieval
and ICSI (26). Therefore, SSR/ICSI should only be offered to
patients where the obstruction is not amenable to surgical
correction or after failed surgery.
The development of SSR/ICSI has revolutionized the man-
agement of infertility caused by azoospermia, with results
Vol. 89, No. 1, January 2008
similar to those obtained using ejaculated sperm (27). This
has led to a number of clinicians (in their understandable
eagerness to help their patients) to adopt a ‘‘see and treat’’
(see azoospermia and do SSR) policy. However, concern
has been raised about the unchecked speed of development
and widespread application of treatments for azoospermia
(28), and clinicians have been sued for using expensive and
invasive treatments for male infertility, albeit successfully,
instead of simpler and potentially useful treatments (29).
There is a need when managing azoospermia to watch out
for the potential pitfalls outlined in this article.
REFERENCES
1. Irvine DS. Epidemiology and aetiology of male infertility. Hum Reprod
1998;13(Suppl 1):33–44.
2. Rowe PJ, Comhaire FH, Hargreave TB, Mellows HJ. WHO manual for
the standardized investigation and diagnosis of the infertile couple.
Cambridge: Cambridge University Press, 1993.
3. Sharif K. Advances in the treatment of male factor infertility. In:
Bonnar J, ed. Recent advances in obstetrics & gynaecology. Vol 21.
Edinburgh: Churchill Livingston, 2001:141–59.
4. Neuwinger J, Behre HM, Nieschlag E. External quality control in the
andrology laboratory: an experimental multicenter trial. Fertil Steril
1990;54:308–14.
5. Jaffe TM, Kim ED, Hoekstra TH, Lipshultz LI. Sperm pellet analysis:
a technique to detect the presence of sperm in men considered to have
azoospermia by routine semen analysis. J Urol 1998;159:1548–50.
6. Greco E, Iacobelli M, Rienzi L, Ubaldi F, Ferrero S, Tesarik J. Reduction
of the incidence of sperm DNA fragmentation by oral antioxidant treat-
ment. J Androl 2005;26:349–53.
7. The Practice Committee of the American Society for Reproductive Med-
icine. The clinical utility of sperm DNA integrity testing. Fertil Steril
2006;86(Suppl 4):S35–7.
8. Finkel DM, Phillips JL, Snyder PJ. Stimulation of spermatogenesis by
gonadotropins in men with hypogonadotropic hypogonadism. N Engl
J Med 1985;313:651–5.
9. Kliesch S, Behre HM, Nieschlag E. High efficacy of gonadotropin or
pulsatile gonadotropin-releasing hormone treatment in hypogonado-
tropic hypogonadal men. Eur J Endocrinol 1994;131:347–54.
10. Nachtigall LB, Boepple PA, Pralong FP, Crowley FW Jr. Adult-onset
idiopathic hypogonadotropic hypogonadism—a treatable form of male
infertility. N Engl J Med 1997;336:410–5.
11. Kamischke A, Nieschlag E. Treatment of retrograde ejaculation and ane-
jaculation. Hum Reprod Update 1999;5:448–74.
12. Mak V, Jarvi KA. The genetics of male infertility. J Urol 1996;156:
1245–56.
13. Rucker GB, Mielnik A, King P, Goldstein M, Schlegel PN. Preoperative
screening for genetic abnormalities in men with nonobstructive azoo-
Fertility and Sterility#
spermia before testicular sperm extraction [see comments]. J Urol
1998;160:2068–71.
14. Kim ED, Bischoff FZ, Lipshultz LI, Lamb DJ. Genetic concerns for the
subfertile male in the era of ICSI. Prenat Diagn 1998;18:1349–65.
15. Donat R, McNeill AS, Fitzpatrick DR, Hargreave TB. The incidence of
cystic fibrosis gene mutations in patients with congenital bilateral
absence of the vas deferens in Scotland. Br J Urol 1997;79:74–7.
16. Costes B, Girodon E, Ghanem N, Flori E, Jardin A, Soufir JC, et al.
Frequent occurrence of the CFTR intron 8 (TG)n 5T allele in men with
congenital bilateral absence of the vas deferens. Eur J Hum Genet
1995;3:285–93.
17. Fedder J, Gabrieslen A, Humaidan P, Erb K, Ernst E, Loft A. Malforma-
tion rate and sex ratio in 412 children conceived with epididymal or
testicular sperm. Hum Reprod 2007;22:1080–5.
18. Schulze W, Thoms F, Knuth UA. Testicular sperm extraction: compre-
hensive analysis with simultaneously performed histology in 1418 biop-
sies from 766 subfertile men. Hum Reprod 1999;14(Suppl 1):82–96.
19. Tournaye H. Use of testicular sperm for the treatment of male infertility.
In: Van Steirteghem A, Tourney H, Devroey P, eds. Male infertility. Bail-
liere’s clinical obstetrics and gynaecology. Vol. 11. London: Bailliere
Tindall, 1997:753–62.
20. Silber SJ, Nagy Z, Devroey P, Tournaye H, Steirteghem V, C. A. Distri-
bution of spermatogenesis in the testicles of azoospermic men: the pres-
ence or absence of spermatids in the testes of men with germinal failure
[published erratum appears in Hum Reprod 1998;13: 780]. Hum Reprod
1997;12:2422–28.
21. Silber SJ, Verheyen G, Goossens A, Devroey P, Steirteghem V, C A. Tes-
ticular sperm distribution in azoospermia. Fertil Steril 1998;70
(Suppl):S197.
22. Plas E, Riedl CR, Engelhardt PF, Muhlbauer H, Pfluger H. Unilateral or
bilateral testicular biopsy in the era of intracytoplasmic sperm injection.
J Urol 1999;162:2010–3.
23. Friedler S, Raziel A, Strassburger D, Komarovsky D, Ron-El R. Intracy-
toplasmic injection of fresh and cryopreserved testicular spermatozoa in
patients with nonobstructive azoospermia—a comparative study. Fertil
Steril 1997;68:892–7.
24. Schlegel PN, Su LM. Physiological consequences of testicular sperm
extraction [see comments]. Hum Reprod 1997;12:1688–92.
25. Amer M, Haggar SE, Moustafa T, Abd El-Naser T, Zohdy W. Testicular
sperm extraction: impact of testicular histology on outcome, number of
biopsies to be performed and optimal time for repetition. Hum Reprod
1999;14:3030–4.
26. Wolter S, Neubauer S, Heidenreich A. Vasovasostomy versus MESA/
TESE combined with ICSI—a cost–benefit analysis. J Urol 1999;161:
312.
27. Tarlatzis BC, Bili H. Survey on intracytoplasmic sperm injection: report
from the ESHRE ICSI Task Force. European Society of Human Repro-
duction and Embryology. Hum Reprod 1998;13(Suppl 1):165–77.
28. Steele EK, Lewis SE, McClure N. Science versus clinical adventurism in
treatment of azoospermia [comment]. Lancet 1999;353:516–7.
29. Jaquire AM. Obstructive azoospermia: a study of 102 patients. Clin
Reprod Fertil 1986;3:21–36.
19